Study of DNA data shows potential for wrongful convictions

A first peek into Virginia's post-conviction DNA project data shows a potential wrongful conviction rate of 6 percent in the decade and a half before DNA testing was widely available.

The preliminary figure from the Urban Institute, which is studying the results, roughly matches the exoneration rate found in 2005 when testing in a small sample of cases cleared two men of rapes and prompted Virginia's full, groundbreaking project.

The 6-year-old effort aimed at clearing innocent people was made possible by a trove of biological evidence samples discovered stored in Virginia Department of Forensic Science case files dating from 1973 through 1988.

John Roman, a senior fellow in the Justice Policy Center at the Urban Institute, said institute researchers reviewing 638 Virginia cases have identified 37 "that might support exoneration and that certainly support further investigation."

Roman said it is hoped the study will indicate how many people were wrongfully convicted of serious crimes from 1973 through 1988.

"I don't know how far we can go down that road," Roman said, "but this is probably the best attempt to get at that number that anybody's ever had. The Justice Department made a really substantial investment in this."

He said authorities today have a huge advantage in preventing mistakes because DNA testing is now available to investigators instead of the less discriminating blood typing done by forensic serologists years ago.

The institute's figures were presented for peer review at a meeting of the American Society of Criminologists in November. Roman stresses that the numbers could change as more research and trips to courthouses across the state continue.

* * * * *

After it was discovered that some state forensic serologists had saved blood and semen samples in their old case files and DNA testing of the material cleared three men of rapes, then-Gov. Mark R. Warner ordered testing in 31 sample cases.

When that testing exonerated an additional two men convicted of rape who had not even sought testing, Warner ordered testing in all appropriate cases the state forensic lab could find.

The department searched 534,000 old case files looking for ones that held samples of biological evidence — primarily blood and semen — and that also resulted in felony convictions. The vast majority of files had no biological evidence samples.

The testing is now virtually completed in the Virginia project, and the Department of Forensic Science reports that in 99 cases the convicted person did not contribute DNA identified by testing.

However, experts say the absence of their DNA alone may have no bearing on innocence and does not mean they were wrongfully convicted.

The Department of Forensic Science's job was to conduct the testing, not to investigate the crimes or the legal significance of the findings.

The state lab has turned the test results in the cases over to authorities where the crimes occurred for appropriate action, if any is warranted.

Efforts are under way to notify all people in whose case files the department found biological evidence. But most of the 99 convicted people whose DNA was not identified have yet to be notified.

Aside from three additional clearances since 2005, prosecutors and law-enforcement agencies apparently have not said anything about the other exclusion cases that may have been forwarded to them by the lab since 2005.

The Virginia project was largely supported by a $4.5 million grant from the Justice Department's National Institute of Justice that paid for testing in most of the cases, though a number of non-NIJ-funded cases also were tested by the lab.

The NIJ grant required Virginia to share the results of the tests it funded with the Urban Institute, which was to use them — and those from a smaller project in Arizona — to try to determine the rate of error in convictions for crimes such as murder, rape and robbery.

Of the 638 cases Virginia sent to the Urban Institute, 419 were classified as "indeterminate" for a variety of reasons that include there being no useable DNA results or the lack of a DNA profile from the convicted person or the victim to compare results with.

Researchers reported in November that their preliminary numbers show that testing in 149 cases identified the convicted person's DNA profile or partial profile and appeared to further implicate the convicted person.

The institute also identified 70 cases in which the convicted person's DNA was not among the DNA identified by testing. Researchers believe 33 of those cases did not appear to support a claim of innocence while 37 did and warranted further investigation.

"We're going to all the courthouses all around the state trying to dig up whatever information we can get about the trials and the events and the facts," Roman said. "We've gone to about a third of the courthouses we plan to get to."

Roman said that once available information is gathered about the potential exonerations it will be presented to an expert prosecutor, a defense lawyer and a former judge for evaluation.

"At some point there's a level of subjectivity that's unavoidable," he said. "We will have a draft to the Justice Department in June and then we should have the final results in September."

* * * * *

Samuel R. Gross, a professor at the University of Michigan Law School and a former criminal-defense lawyer, attended the presentation made by the Urban Institute in November and said that "this is a very big surprise."

"I would have guessed an error rate of 1 or 2 percent. Six percent is surprisingly high," Gross said.

"Beyond that there's this truly troubling fact that somebody now has (exculpatory) information about three dozen people who were convicted … and we have heard nothing about these cases. They haven't resulted in exonerations with the possible exception of the two or three cases that have become public," Gross said.

But Fairfax County Commonwealth's Attorney Raymond F. Morrogh, a member of the Virginia Board of Forensic Science, complained at the board's meeting last week that the board had not been sent the Urban Institute's preliminary findings.

"It troubles me that they're out talking about what we did without sending us their thoughts so our scientists can look at them," he said, requesting that the department obtain a copy.

"I'd like a chance, personally, to evaluate what they're saying," Morrogh said.

According to the Urban Institute, the Justice Policy Center "conducts nonpartisan research and evaluation designed to improve justice and public safety policies and practices at the national, state and local level."